=== maclin1 is now known as maclin === maclin1 is now known as maclin [08:25] tsdgeos, hey, I thought of something re: huge background, could we put it through the thumbnailer so it caches the scaled down version? [08:26] Saviq: we could, but that would not help with smooth resizing i guess [08:26] tsdgeos, yeah it would, I'm only after not resizing from the 4K bitmap every time we load [08:27] tsdgeos, like you did in the branch, just instead of loading directly, put it through thumbnailer [08:27] this way the next time we load it, we should have a cached, scaled down version [08:34] Saviq: ok, i was still waking up :D [08:34] nw :) [08:34] Saviq: i guess it wold work yes [08:38] Saviq: want me to alter the MR or stack it on top in a different MR? [08:39] tsdgeos, same MR is fine [08:42] k [09:01] Saviq: pushed, please make sure the commit log i made makes sense :D [09:01] tsdgeos, tak [09:02] tsdgeos, hmm thumbnailer can't deal with file:// urls? [09:04] tsdgeos, according to https://developer.ubuntu.com/en/apps/qml/tutorials/use-ubuntu-thumbnailer/ you shouldn't need the .substr() [09:04] hmmm [09:04] it didn't work i think last time i tried [09:04] let me see [09:05] oh it does work [09:05] weird i must have been doing something wrong [09:05] yay [09:05] good catch [09:05] it did indeed look ugly [09:28] how could i try the new "ubuntu personal" unity8 desktop? using an iso or using a lxd session? [09:56] dkessel, you need to install unity8-desktop-session-mir and select the "Unity 8" session in greeter [09:56] no other way today I'm afraid [09:59] ok thanks Saviq [10:03] lol [10:03] QWARN : tst_DragHandle::dragThreshold_horizontal(rotated 90) file:///usr/lib/x86_64-linux-gnu/qt5/qml/Ubuntu/Components/Themes/Ambiance/1.3/ButtonStyle.qml:45: TypeError: Property 'gu' of object function() { [code] } is not a function [10:03] the line is [10:03] implicitHeight: units.gu(4) [10:14] what is "units"? [10:23] you know the typical units.gu thing we use everywhere [10:23] yeah I know, but what it is in this case [10:24] sounds like the context property didn't get set proper (/me hates context properties btw, they're too "magically there" :P) [10:27] Saviq: it's the problem we're finding with Qt 5.5 and the SDK having statics [10:27] i'm refining the problem to give zsombi a reproducible case and stumbled upon it [10:27] more fun than something that i wanted really to fix [10:27] ack [10:27] :D [10:36] https://bugs.launchpad.net/ubuntu/+source/ubuntu-ui-toolkit/+bug/1527546 is the crash/bg [10:36] Ubuntu bug 1527546 in ubuntu-ui-toolkit (Ubuntu) "Crash when having multiple QQuickViews in a row" [Undecided,New] [10:42] dednick: you here? [10:45] Saviq: are our xenial silos building fine? [10:45] xavigarcia is having a trouble with one giving him https://launchpadlibrarian.net/230433259/buildlog_ubuntu-xenial-armhf.unity8_8.11%2B16.04.20151218.1-0ubuntu1_BUILDING.txt.gz [10:45] tsdgeos, just finding out now, yesterday there was a problem with some other migration [10:45] that if i read correclty seems a dependncy problem? [10:46] tsdgeos, yeah, if that's current then we're stuck, proposed has some migration issues [10:46] * Saviq will investigate in a mo [10:46] oka [10:46] was hoping it'd be gone by now [10:46] not sure how recent it is, let me ask [10:47] Saviq: seems it's from today :/ [10:47] but yeah if you can push some buttons it'd be great [10:49] tsdgeos: i am [10:49] dednick: k, xavigarcia was looking for you and you don't seem to be on the canonical irc [10:49] tsdgeos: i can't get on either [10:49] not for like 2 days now. [10:50] oh [10:51] tsdgeos: ah. removing dns got me back on === tsdgeos_ is now known as tsdgeos [11:05] tsdgeos, hah, http://pastebin.ubuntu.com/14087111/ [11:05] boooo [11:06] so we did not rebuild that? [11:06] i mean how did that happen? [11:06] we had a landing after 5.5.1, no? [11:08] yeah, I think it's an explicit dep [11:09] tsdgeos, bug #1527544 [11:09] bug 1527544 in oxide-qt (Ubuntu) "liboxideqt-qmlplugin 1.11.3-0ubuntu1 in xenial-proposed not installable" [Critical,In progress] https://launchpad.net/bugs/1527544 [11:10] wonder how did that happen [11:12] ah that's not going through silos === boiko_ is now known as boiko [12:09] tsdgeos, pstolowski did you try pawel scope with silo 54? crashes for me when clicking filters [12:14] cimi, what crashes? the shell? [12:14] the scope [12:14] on desktop too [12:14] when I click on the filters [12:15] yaeh the scope doesn't work [12:15] cimi, hmm apparently something broke since i last checked. will take a look [12:15] i think it needs a rebuild or something [12:15] it has that behaviour of blocking the dash [12:15] becaue it takes too much to start/crashes [12:16] terminate called after throwing an instance of 'unity::LogicException' [12:16] what(): unity::LogicException: Variant does not contain a double value: [12:16] boost::bad_get: failed value get using boost::get [12:16] oh ok then other stuff... [12:16] noticed now the log continues [12:49] tsdgeos, re the crash of filters test scope - do you has hasStartValue() before calling startValue() on rangeinput filters as agreed before? [12:49] s/has/call/ [13:01] in any case i'm changing it to issue a warning and return 0 instead of throwing. not ideal, but better than aborting === dandrader is now known as dandrader|afk === alan_g is now known as alan_g|lunch [13:07] pstolowski: let me check [13:08] pstolowski: but anyway, yeah never throw :D [13:09] pstolowski: that'd be in the slider value, no? [13:09] tsdgeos, no, it's the simple range input filter with to input boxes [13:09] ah right [13:09] tsdgeos, so no value is perfectly correct [13:10] yeah [13:10] so i do use it and do not use it [13:10] :) [13:11] you better change it [13:11] it makes the qml/js life easier [13:11] ok, just make sure you don't show the value if has*Value is false [13:14] pstolowski: so what i was saying [13:14] is that i have code that does http://paste.ubuntu.com/14087668/ [13:14] so it is not used, but it is accessed [13:14] since it makes it easier to have a function that does the check [13:14] instead of copying the code 4 times [13:15] tsdgeos, ah, i see === dandrader|afk is now known as dandrader [13:36] cimi: you're doing https://code.launchpad.net/~stolowski/unity8/shareData-rename/+merge/280056, right? [14:14] mzanetti, revision 2094 broke some code. you removed priv.foregroundMaximizedAppIdIndex but it's still being referenced in a couple of places [14:14] mzanetti, in DesktopStage.qml [14:15] uh [14:15] kk, will fix === alan_g|lunch is now known as alan_g [14:37] cimi, tsdgeos i fixed the crash in silo 54 === dandrader is now known as dandrader|lunch [14:39] :) [14:41] tsdgeos, yes [15:06] ltinkl, re: when to update .po/.pot files, I realized that you might if you're aware of one more thing in Ubuntu - langpacks - which are generated out-of-band from the normal release process and updated periodically [15:07] Saviq, I see, thx [15:09] ltinkl, as you can see unity8 packages don't actually ship the .mo files, they're stripped off of the packages and langpacks are generated as needed: http://packages.ubuntu.com/search?searchon=contents&keywords=unity8.mo&mode=exactfilename&suite=xenial&arch=any [15:10] ltinkl, https://wiki.ubuntu.com/UbuntuDevelopment/Internationalisation/Packaging#Language_Packs [15:11] actually now I look at it am not so sure langpacks are updated outside of package releases :P [15:13] Saviq, ye it doesn't mention that; how does it fit with our different release process? [15:15] ltinkl, at the very least langpacks are generated before every OTA [15:15] Saviq, yea, at least that :) so rc-proposed is likely left out, right? [15:15] ltinkl, oh yeah, that's for sure ;) [15:16] OTOH the latest langpack is from 1216 [15:16] Saviq, ugh that doesn't look good [15:17] Saviq, 2014-08-27 really? [15:17] ltinkl, 12.16 [15:17] 2015.12.16 [15:18] ah :) I thought revision number [15:18] or 1216, the true medieval language pack [15:19] ltinkl, they seem to be updated every week for "in-flight" archives https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/stable-phone-overlay/+packages?field.name_filter=language-pack-touch-pl&field.status_filter=&field.series_filter=vivid [15:20] Saviq, in-flight? [15:21] ltinkl, vivid overlay, xenial (not released) [15:21] tsdgeos, have two failures in silo 31 http://pastebin.ubuntu.com/14088316/, anything springs to mind? [15:21] that's on vivid [15:22] hmmm === dandrader|lunch is now known as dandrader [15:34] tsdgeos, pstolowski the range filter makes it crash [15:35] when you set min and max [15:35] cimi: i think pstolowski was fixing that? [15:35] that's another one then i wasnt aware [15:35] no, looks like something else [15:35] try setting a min and max value, it crashes [15:36] cimi, ignore range filter for now. it's the first time i actually see it on the screen after implementing it. my test scope doesn't even honor it yet [15:36] ok so I will just review the optionselector of albert [15:51] Saviq: can't see anything in silo 31 that would cause these tbh [15:52] Saviq: is there a branch with all the branches merged or do i create one for myself? [15:52] tsdgeos, lp:~ci-train-bot/unity8/ has them all [15:53] tsdgeos, lp:~ci-train-bot/unity8/unity8-ubuntu-xenial-landing-031 in particular [15:53] tsdgeos, could be flaky, is all, I will run adt again [15:53] will try a local build as well [15:56] Saviq: also adt has different/more failures than qmluitests sometimes because we don't "split/organize" things for it (as we seldom run it) [15:56] so that may be a reason too [15:56] tsdgeos, might be indeed === dandrader is now known as dandrader|afk [16:06] tsdgeos, refine your results doesnt work for me too [16:07] tsdgeos, did you try the test scope by pawel? [16:07] cimi: i did, like 3 months ago [16:08] cooool :D [16:08] cimi: what "doesn't work2? [16:08] I set two optionselector filters, I press "refine your result" reset button [16:08] and nothing happens [16:09] "Refine results" is a label [16:09] the "Reset" button is not linked to it [16:09] but the reset next to it? [16:09] it's a UI fail if you think they are :D [16:09] yes there's a reset button next to it [16:10] yeah is UI fail indeed [16:10] but the reset button, that looks like a label and not a button, doesnt click [16:10] i'mk not sure if he has implement the restet function [16:11] it does, when running make tryDash and clicking on it i can clearly see the debug line printed [16:11] ah ok [16:11] so the bug can be [16:11] a) reset is not implement on his side [16:12] I see your code just does scopeView.scope.resetFilters() [16:12] so it belongs to pawel [16:12] b) the reset implementation on his side breaks on the UI level when propagating the changes [16:12] anchors seems fine too [16:12] the AButton is on top of the label so... [16:12] ok [16:13] Saviq: i've looped dashheadertest on that branch and doesn't seem to fail [16:14] tsdgeos, I'm gonna approve those two optionselector branches, they seem to work apart this unimplemented reset [16:14] tsdgeos, the ranges one are not testable on a real scope, they crash, so I'd hold on on them [16:14] yeah, well he's away for the next week [16:15] so i guess we can continue next year :) [16:15] tsdgeos, it's not a good reason not to approve option selector :) [16:15] right [16:15] that would be nice [16:15] lands 4 branches [16:16] not sure how doable it is though [16:16] otoh we can land all the branches too [16:16] since no scope uses those filters [16:16] so even if there's some issues it won't matter [16:16] and the tests on our side "kind of prove" they "kind of work" [16:16] well, option selector works and I tested, I would not approve the ranges or you would too? [16:17] not that we can't land anything yet anyway since the oxide-qml thing :D [16:17] * tsdgeos gets a cookie [16:22] tsdgeos, ack [16:23] tsdgeos, approved \o/ [16:25] :) [16:29] cimi: you approving the 2 parents? [16:30] yeah [16:30] big xmas sales! [16:31] :D [16:31] approving branches like santa is giving away gifts [16:37] :) === dandrader|afk is now known as dandrader [16:43] tsdgeos: what's wrong with oxide-qml? I've discovered that I couldn't start webbrowser-app on xenial anymore, does it just need a rebuild against latest qt or is it something else? [16:44] (since you are meintionning it...) [16:44] davidcalle: yeah needs a rebuild it seems [16:45] ok, ty [16:45] davidcalle: https://bugs.launchpad.net/ubuntu/+source/oxide-qt/+bug/1527544 [16:45] Ubuntu bug 1527544 in oxide-qt (Ubuntu) "liboxideqt-qmlplugin 1.11.3-0ubuntu1 in xenial-proposed not installable" [Critical,In progress] [16:53] davidcalle, note this problem's only in proposed, so if you don't have proposed enabled, your issue's something else [17:37] ChrisTownsend, still need that update on the best way to run Unity8 atm. Can you spread some love on https://wiki.ubuntu.com/Unity8inLXC and https://wiki.ubuntu.com/Unity8Desktop ? [17:46] balloons: Oh, right, sorry, forgot about those. === dandrader is now known as dandrader|afk [18:06] balloons: Ok, I've updated those pages. === alan_g is now known as alan_g|EOY [18:07] ChrisTownsend, lovely thank you! :-) [18:07] balloons: You're welcome. === dandrader|afk is now known as dandrader